Diversity of risk of mother-to-child HIV-1 transmission according to feeding practices, CD4 cell count, and haemoglobin concentration in a South African cohort.

OBJECTIVE: To estimate the probabilities of intrapartum and postpartum HIV-1 mother-to-child-transmission according to various feeding practices (formula feeding, exclusive breastfeeding, or mixed-feeding) and to other mother and infant covariates.
METHODS: We used the promotion time model extended to multiple exposures to study the probability of infection attributable to each transmission occasion. Blood samples from 551 infants from Durban (South Africa) born to HIV-1 positive untreated mothers between 1995 and 1998 were sequentially tested until 15 months.
RESULTS: The probability of infection attributable to in utero and intrapartum transmission was 21.88% (18.71-25.20) and was not significantly associated with the feeding practice. The probability of infection attributable to postnatal transmission through exclusive breastfeeding was negligible -0.7% (0-2.5) for 6 months of exclusive breastfeeding-- in comparison with that observed with mixed-feeding -6.15% (3.16-9.57) for 6 months of mixed-feeding. Maternal CD4 cell count and gestational age were significant predictors of intrapartum transmission probability while maternal CD4 cell count and maternal haemoglobin concentration were significant predictors of postpartum transmission probability.
CONCLUSION: Decisions about appropriate infant feeding practices should take into account the difference in postpartum transmission risk between exclusive and mixed-feeding. Mixed-feeding should be all the more avoided that mothers have poor immunological statuses and low haemoglobin concentrations.